Impact of blood pressure lowering,cholesterol lowering and their combination in Asians and non-Asians in those without cardiovascular disease: an analysis of the HOPE 3 study

Abstract

Background and design

There are limited data on the effects of blood pressure and cholesterol lowering in Asians at intermediate risk and no cardiovascular disease. We report an analysis of the effects of blood pressure and cholesterol lowering in Asians enrolled in the Heart Outcomes Prevention Evaluation 3 (HOPE 3) trial.

Methods

We randomly assigned 6241 Asians and 6464 non-Asians at intermediate risk without cardiovascular disease to candesartan 16 mg/hydrochlorothiazide 12.5 mg or placebo and rosuvastatin 10 mg or placebo. The first co-primary outcome was a composite of cardiovascular disease death, myocardial infarction and stroke. The second co-primary outcome additionally included heart failure, cardiac arrest and revascularisation. Median follow-up was 5.6 years.

Results

Reduction in systolic blood pressure was less among Asians (4.3 vs. 7.7 mmHg for non-Asians, P < 0.0001) mainly due to a lesser effect in Chinese (2.1 mmHg) than in other Asians (7.3 mmHg), reduction in the latter being similar to non-Asians. The effect on the composite outcomes was similar, with no significant benefits from blood pressure lowering for either Asians (Chinese or non-Chinese) or non-Asians. Rosuvastatin reduced low-density lipoprotein cholesterol to a lesser degree in Asians (0.49 mmol/L (–19.1 mg/dL) compared with non-Asians 0.95 mmol/L (−36.7 mg/dL), P_interaction < 0.0004). Yet both groups had similar reductions in the two co-primary outcomes. There was no increase in permanent medication discontinuation due to muscle-related symptoms in either group. There was an excess in new diabetes in non-Asians (4.70% rosuvastatin, 3.52% placebo, P = 0.025) but not in Asians (3.02% rosuvastatin, 4.04% placebo, P = 0.0342), P_interaction = 0021.

Conclusions

Candesartan/hydrochlorothiazide had fewer effects in reducing blood pressure in Chinese and rosuvastatin reduced low-density lipoprotein cholesterol to a lesser extent in Asians compared with non-Asians. There was no overall reduction in clinical events with lowering blood pressure in either Asians or non-Asians, whereas there were clear and consistent benefits with lipid lowering in both. Despite extensive analyses, we have no obvious explanation for the observed findings. Future studies need to include larger numbers of individuals from different regions of the world to ensure that the results of trials are applicable globally.

Keywords

Clinical trial primary prevention polypill Asians Chinese

Introduction

Eighty per cent of the global burden of cardiovascular disease (CVD) occurs in low and middle income countries.¹ The recommended strategy for primary prevention^2,3 – identifying people at risk and titrating medications individually to target levels of risk factors – requires multiple visits to a healthcare facility, carries a high burden on staff and patients, increases costs and may reduce adherence. In 2003 Wald and Law⁴ proposed that, if all people older than 55 years or with CVD were to be given a fixed-dose multicomponent pill (with medications to lower blood pressure (BP), low-density lipoprotein (LDL) cholesterol, aspirin and folic acid) myocardial infarction (MI) and stroke could be reduced by 80%, regardless of specific risk factor levels. Although the efficacy and safety of these combination therapies have been studied,^5,6 until recently there have been few data on reduction in clinical events. The Heart Outcomes Prevention Evaluation 3 (HOPE 3) trial^7–9 demonstrated significant reductions in CVD events with rosuvastatin irrespective of base line LDL-cholesterol or other characteristics. However, BP lowering with the combination of candesartan and hydrochlorothiazide did not reduce CVD in the overall population, except in those in the highest third of systolic blood pressure (SBP) (>143 mmHg). The trial was conducted in 21 countries with a large representation from Asia, providing an opportunity to explore the consistency of the effects of both interventions within and outside Asia.

Although about half the world’s population lives in Asia, there are limited data on cholesterol or BP lowering in Asians. Furthermore, there have been concerns that statin use may be associated with more adverse effects in Asians.^10,11 We report here a prespecified analysis from the HOPE 3 trial comparing the efficacy and safety of both study treatments in Asians and non-Asians.

Methods

The study design and main outcomes have previously been published.^7–9 In brief HOPE 3 is a multicentre international double blind, placebo controlled, randomised clinical trial utilising a 2 × 2 design. The goal was to compare cardiovascular event rates using a fixed dose combination of candesartan 16 mg and hydrochlorothiazide 12.5 mg in a single pill (BP-lowering arm), rosuvastatin 10 mg (cholesterol-lowering arm), both or neither in people without CVD but at intermediate risk of a cardiovascular event (approximately 10% over 10 years). Astra Zeneca provided the trial medications. The trial was approved by the ethics committee at each trial centre and by respective regulatory authorities. All participants were given individualised lifestyle advice. Follow-up visits occurred at 6 weeks post-random selection and thereafter every 6 months. At each visit adherence to medication, BP (using an automated Omron model HEM-711DLXCAN), safety and occurrence of clinical events were recorded. Serum lipids were estimated from a fasting sample at baseline in all participants. Measurements were repeated at one year, 3 years and at study end in a sample of study participants in different countries (Table 1).

Table 1.

Central LDL-cholesterol measurement.

	Randomised	Baseline	1 Year	3 Years	End of study
Asian	6241	5309	604	773	1014
Non-Asian	6464	6089	927	772	687
Total	12,705	11,398	1531	1545	1701

LDL: low-density lipoprotein.

The trial had two co-primary outcomes. Co-primary outcome 1 was the composite of death from cardiovascular causes, non-fatal MI or non-fatal stroke; co-primary outcome 2 included the first composite outcome plus resuscitated cardiac arrest, heart failure or revascularisation. Secondary outcomes included individual components of the co-primary outcomes, and the second co-primary outcome plus angina with evidence of ischaemia. The main safety outcomes included myopathy, rhabdomyolysis and any hospitalisation. All cardiovascular events and cases of new diabetes were adjudicated.

For the present analysis participants were classified as Asian based on self-declared ethnicity irrespective of country of recruitment. Classification based on country of recruitment (Asian – China, India, Republic of Korea, Malaysia and Philippines is very similar, see Table 2). Non-Asians included black African, white Caucasian, Latin American and others.

Table 2.

Distribution of subjects (Asian or non-Asian) by country of recruitment or by declared ethnicity.

	By country^a	By self-declared ethnicity^b
Asian	6173	6241
Others	6532	6464
Total randomly assigned	12,705	12,705

China 3677, India 1824, Republic of Korea 14, Malaysia 87, Philippines 571.

Chinese 3691, South Asian 1854, other Asian 696, black African 225, white Caucasian 2546, Latin American 3496, other 197.

Statistical analysis

Comparisons of the effects on BP, lipids and the clinical events of each treatment versus its respective placebo and of the combination versus double placebo were calculated among Asians and non-Asians using Cox proportional hazards models stratified according to the factorial allocation. The hazard ratio (HR) for each treatment versus its placebo was calculated among Asians and non-Asians and was compared using tests of interaction. Analyses were performed in all randomly assigned subjects and also in those in the upper third of SBP. Analyses were done using SAS version 9.4 (SAS Institute Inc., NC, USA) and figures were generated using S-PLUS 8.2 (TIBCO Software Inc., CA, USA).

Results

Between April 2007 and November 2010 12,705 participants from 228 sites in 21 countries were randomly assigned. Of these 6241 were Asian. The median follow-up was 5.6 years. Asians were younger and had lower body mass index (BMI) and waist-to-hip ratios than non-Asians. However, the prevalence of tobacco use and diabetes mellitus were slightly higher among Asians. As a result, risk scores were slightly higher in Asians than non-Asians (Table 3). Asians were also less likely to be taking aspirin or medications for hypertension. The median follow-up for Asian and non-Asian participants was 5.5 (interquartile range (IQR) 5.0–6.0) and 5.9 (IQR 5.3–6.7) years, respectively. The baseline characteristics of Asians and non-Asians by study arm are shown in Tables 4 and 5 and are broadly similar.

Table 3.

Baseline characteristics in Asians and non-Asians.

Characteristic	Asians (n = 6241)	Non-Asians (n = 6464)	P value
Mean age (SD)	64.91 (6.00)	66.49 (6.60)	<0.0001
Gender (male %)	3299 (52.9)	3532 (54.6)	0.0441
Current/recent smoking (%)	2021 (32.4)	1503 (23.3)	<0.0001
Diabetes mellitus	404 (6.5)	335 (5.2)	0.0019
Hypertension (%)	2368 (37.9)	2446 (37.8)	0.9054
Body mass index (SD)	25.65 (4.07)	28.53 (4.98)	<0.0001
Waist to hip ratio (SD)	0.92 (0.07)	0.95 (0.09)	<0.0001
Interheart score (SD)	14.84 (5.13)	14.14 (5.21)	<0.0001
SBP mmHg (SD)	139.84 (14.28)	136.33 (15.03)	<0.0001
DBP mmHg (SD)	82.44 (9.06)	81.33 (9.53)	<0.0001
Total cholesterol, mg/dl (SD)	200.70 (43.78)	202.07 (40.69)	0.0813
LDL-cholesterol, mg/dl (SD)	128.99 (38.07)	126.85 (34.21)	0.0016
HDL-cholesterol, mg/dl (SD)	42.88 (14.15)	46.51 (13.32)	<0.0001
Triglycerides, mg/dl (SD)	148.70 (101.79)	147.82 (84.41)	0.6126
Apo-lipoprotein B, g/l (SD)	0.99 (0.28)	1.05 (0.24)	<0.0001
Apo-lipoprotein A1, g/l (SD)	1.37 (0.30)	1.54 (0.35)	<0.0001
Fasting glucose, mg/dl median (IQR)	95.40 (86.40–108.00)	95.40 (87.00–104.40)	<0.0001
hsCRP median, mg/l (IQR)	1.84 (0.93–3.60)	2.20 (1.11–4.30)	<0.0001
Serum creatinine, mg/dl (SD)	0.86 (0.22)	0.93 (0.20)	<0.0001
Concomitant medications, n (%)
Aspirin	597 (9.6)	796 (12.3)	<0.0001
Beta-blockers	226 (3.6)	794 (12.3)	<0.0001
Calcium antagonists	757 (12.1)	1128 (17.5)	<0.0001
Alpha blockers	20 (0.3)	121 (1.9)	<0.0001
Non-thiazide diuretics	11 (0.2)	54 (0.8)	<0.0001
Any blood pressure-lowering drugs	948 (15.2)	1835 (28.4)	<0.0001
Oral hypoglycaemic agents	177 (2.8)	160 (2.5)	0.2057

DBP: diastolic blood pressure; HDL: high-density lipoprotein; hsCRP: high sensitivity C-reactive protein; IQR: interquartile range; LDL: low-density lipoprotein; SBP: systolic blood pressure; SD: standard deviation.

Table 4.

Baseline characteristics at run in visit by study arm: candesartan.

	Asian		Non-Asian
Characteristics	Candesartan/HCTZ (N = 3118)	Placebo (N = 3123)	Candesartan/ HCTZ (N = 3238)	Placebo (N = 3226)
Mean age (SD), years	64.75 (5.99)	64.88 (6.02)	66.40 (6.62)	66.41 (6.58)
Male, N (%)	1673 (53.7)	1626 (52.1)	1773 (54.8)	1759 (54.5)
Current smoking, N (%)	857 (27.5)	767 (24.6)	482 (14.9)	508 (15.8)
Diabetes mellitus, N (%)	208 (6.7)	196 (6.3)	182 (5.6)	153 (4.7)
Hypertension N (%)	1180 (37.8)	1188 (38.0)	1218 (37.6)	1228 (38.1)
Elevated WHR	2571 (82.5)	2597 (83.2)	2940 (90.8)	2926 (90.7)
BMI mean (SD)	25.61 (4.12)	25.70 (4.03)	28.55 (5.01)	28.51 (4.95)
SBP at run in, mmHg (SD)	139.86 (14.15)	139.82 (14.44)	136.61 (15.16)	135.96 (14.9)
DBP at run in, mmHg (SD)	82.55 (9.14)	82.31 (9.00)	81.38 (9.58)	81.21 (9.45)
Interheart risk score, N (%)	14.87 (5.13)	14.80 (5.14)	14.22 (5.17)	14.05 (5.26)
Framingham risk score, N (%)	23.28 (13.41)	22.68 (13.31)	21.37 (11.99)	21.22 (11.96)
Total cholesterol,^a mg/dL (SD)	200.4 (43.63)	201.0 (43.93)	202.2 (41.74)	201.9 (39.62)
LDL-cholesterol,^a mg/dL (SD)	128.4 (38.25)	129.6 (37.88)	126.6 (34.96)	127.1 (33.45)
HDL-cholesterol,^a mg/dL (SD)	43.1 (14.23)	42.7 (14.18)	46.5 (13.44)	46.6 (13.21)
Triglycerides,^a mg/dL(SD)	149.6 (105.1)	147.8 (98.39)	149.9 (88.53)	145.7 (80.04)
Apo B,^a g/l (SD)	0.99 (0.28)	0.99 (0.28)	1.05 (0.21)	1.05 (0.24)
Apo A1,^a g/l (SD)	1.37 (0.31)	1.36 (0.30)	1.54 (0.35)	1.54 (0.35)
CRP,^a mg/l	3.61 (6.64)	3.21 (5.47)	3.80 (5.71)	3.79 (5.79)

Percentages were calculated out of data available.

Apo: apolipoprotein; BMI: body mass index; DBP: diastolic blood pressure; HCTZ: hydrochlorothiazide; HDL: high-density lipoprotein; CRP: C-reactive protein; LDL: low-density lipoprotein; SBP: systolic blood pressure; SD: standard deviation; WHR: waist to hip ratio.

Central laboratory blood results.

Table 5.

Baseline characteristics at run in visit by study arm: rosuvastatin.

	Asian		Non-Asian
Characteristics	Rosuvastatin (N = 3122)	Placebo (N = 3119)	Rosuvastatin (N = 3239)	Placebo (N = 3225)
Mean age (SD), years	64.88 (5.95)	64.75 (6.06)	66.42 (6.69)	66.39 (6.51)
Male, N (%)	1627 (52.1)	1672 (53.6)	1783 (55.0)	1749 (54.2)
Current smoking, N (%)	815 (26.1)	809 (25.9)	486 (15.0)	504 (15.6)
Diabetes mellitus, N (%)	204 (6.5)	200 (6.4)	174 (5.4)	161 (5.0)
Hypertension, N (%)	1180 (37.8)	1188 (38.1)	1223 (37.8)	1223 (37.9)
Elevated WHR	2591 (83.0)	2577 (82.6)	2949 (91.0)	2917 (90.4)
BMI mean (SD)	25.77 (4.16)	25.54 (3.98)	28.50 (4.96)	28.56 (5.00)
SBP at run in, mmHg (SD)	139.76 (14.46)	139.92 (14.13)	136.35 (15.19)	136.22 (14.88)
DBP at run in, mmHg (SD)	82.40 (9.01)	82.46 (9.12)	81.29 (9.69)	81.30 (9.34)
Interheart risk score, N (%)	14.88 (5.08)	14.80 (5.19)	14.22 (5.29)	14.05 (5.13)
Framingham risk score, N (%)	22.91 (13.53)	23.05 (13.19)	21.46 (12.21)	21.13 (11.74)
Total cholesterol,^a mg/dL (SD)	200.7 (43.95)	200.8 (43.61)	202.3 (41.34)	201.8 (40.05)
LDL-cholesterol,^a mg/dL (SD)	128.4 (38.17)	129.2 (37.96)	126.9 (34.24)	126.8 (34.18)
HDL-cholesterol,^a mg/dL (SD)	43.1 (14.37)	42.6 (13.93)	46.1 (13.34)	46.9 (13.29)
Triglycerides,^a mg/dL(SD)	147.0 (89.62)	148.3 (100.57)	149.1 (103.2)	150.5 (91.88)
Apo B,^a g/l (SD)	0.99 (0.28)	0.98 (0.28)	1.06 (0.24)	1.05 (0.23)
Apo A1,^a g/l (SD)	1.37 (0.31)	1.36 (0.29)	1.53 (0.35)	1.55 (0.35)
CRP,^a mg/l	3.38 (5.70)	3.45 (6.45)	3.85 (5.92)	3.74 (5.57)

Percentages were calculated out of data available.

Apo: apolipoprotein; BMI: body mass index; DBP: diastolic blood pressure; HDL: high-density lipoprotein; CRP: C-reactive protein; LDL: low-density lipoprotein; SBP: systolic blood pressure; SD: standard deviation; WHR: waist to hip ratio.

Central laboratory blood results.

Permanent discontinuation of both study medications was lower among Asians compared with non-Asians (Table 6). By the end of the study in the candesartan/hydrochlorothiazide arm, 14.6% of Asians had stopped study medication compared with 12.6% in the placebo arm. For non-Asians the corresponding figures were 31.7% and 35.9%. In the rosuvastatin arm, 13.1% of Asians had stopped study medications compared with 13.4% in the placebo arm. Among non-Asians the corresponding numbers were 32.2% and 36.9%.

Table 6.

Adherence to study medications.

	Asians				Non-Asians
	Eligible	On study medication	Eligible	On study medication	Eligible	On study medication	Eligible	On study medication
Visit	Candesartan/ hydrochlorothiazide		Placebo		Candesartan/ hydrochlorothiazide		Placebo
Randomised	3118		3123		3238		3226
12 Months	3104	2858 (92.1%)	3107	2877 (92.6%)	3210	2709 (84.4%)	3199	2668 (83.4%)
24 Months	3084	2832 (91.8%)	3088	2854 (92.4%)	3183	2542 (79.9%)	3174	2505 (78.9%)
EOS	2972	2537(85.4%)	2981	2605 (87.4%)	3018	2062 (68.3%)	3004	1925 (64.1%)
	Rosuvastatin		Placebo		Rosuvastatin		Placebo
Randomised	3122		3119		3239		3225
12 Months	3110	2858 (91.9%)	3101	2890 (93.2%)	3217	2707 (84.1%)	3192	2638 (82.6%)
24 Months	3092	2841 (91.9%)	3080	2844 (92.3%)	3194	2543 (79.6%)	3163	2468 (78.0%)
EOS	2988	2596 (86.9%)	2965	2568 (86.6%)	3026	2053 (67.8%)	2996	1889 (63.1%)

EOS: end of study.

BP-lowering arm

Asians had less reduction in SBP in the active BP-lowering arm (4.3 ± 12.62 mmHg) compared to non-Asians (7.7 ± 13.33 mmHg), P_interaction < 0.0001) (Figure 1 and Table 7). Among Asians the SBP lowering was significantly more modest among Chinese (difference in SBP of –2.1 ± 11.17 and DBP of –1.1 ± 7.59 between active and control) compared with non-Chinese Asians (difference of SBP –7.3 ± 14.08 and DBP of –3.4 ± 8.45 between active and control (Table 8). This lower reduction in BP among Chinese was not due either to the increased use of open label BP-lowering medications or drug discontinuation (Tables 9 and 10).

Figure 1.

Systolic and diastolic blood pressure by candesartan/hydrochlorothiazide and placebo in Asians and non-Asians.

Table 7.

Reductions in blood pressure, lipids and hs-CRP in Asians and non-Asians.

	Asians		Non-Asians
	Candesartan/HCTZ	Placebo	Candesartan/HCTZ	Placebo	P_interaction Asian vs. non-Asians
	N = 3118	N = 3123	N = 3238	N = 3226	P_interaction Asian vs. non-Asians
Blood pressure-lowering arm
SBP, mmHg (SD)
Baseline	139.9 (14.1)	139.8 (14.4)	136.6 (15.1)	136.0 (14.9)
Mean differences	–10.0 (12.5)	–5.7 (12.7)^a	–10.1(13.7)	–2.4 (12.9)^a	<0.0001
DBP, mmHg (SD)
Baseline	82.5 (9.1)	82.3 (9.0)	81.4 (9.6)	81.3 (9.5)
Mean differences	–5.5 (8.1)	–3.4 (7.9)^a	–5.9 (8.4)	–1.9 (7.7)^a	<0.0001
	Rosuvastatin	Placebo	Rosuvastatin	Placebo
	N = 3122	N = 3119	N = 3239	N = 3225
Cholesterol-lowering arm
Total cholesterol, mg/dL (SD)
Baseline	206.82 (47.16)	207.09 (41.94)	206.88 (33.75)	205.60(34.38)
Mean differences	–27.44 (47.64)	–8.59 (42.05)^a	–44.32 (42.91)	–5.26 (37.45)^a	0.0002
LDL-cholesterol, mg/dL (SD)
Baseline	133.86 (39.55)	133.51 (34.82)	131.43 (30.07)	129.12 (29.75)
Mean differences	–29.42 (43.72)	–10.30 (35.55)^a	–42.95 (39.11)	–6.26 (34.48)^a	0.0004
HDL-cholesterol, mg/dL (SD)
Baseline	45.89 (12.55)	46.55 (11.72)	46.22 (11.76)	47.49 (13.26)
Mean differences	2.81 (13.56)	1.60 (11.17)	3.87 (9.46)	1.47 (8.31)^a	0.3785
Triglycerides, mg/dL (SD)
Baseline	137.31 (78.09)	140.43 (117.72)	151.88 (86.15)	147.47 (70.82)
Mean differences	–4.52 (89.88)	–2.38 (112.76)	–29.08 (80.08)	3.53 (84.13)^a	0.0096
Apo A, g/L (SD)
Baseline	1.46 (0.29)	1.47 (0.26)	1.55 (0.34)	1.56 (0.32)
Mean differences	0.10 (0.29)	0.11 (0.30)	–0.07 (0.29)	–0.11 (0.25)	0.1881
Apo B, g/L (SD)
Baseline	0.98 (0.29)	0.96 (0.26)	1.10 (0.22)	1.08 (0.22)
Mean differences	–0.03 (0.33)	0.13 (0.30)^a	–0.26 (0.25)	–0.03 (0.21)^a	0.0409
hsCRP, mg/L (SD)
Baseline	2.65 (5.04)	2.21 (3.33)	3.46 (4.04)	3.48 (4.47)
Mean differences	–0.30 (5.60)	0.23 (6.30)	–0.23 (5.06)	0.26 (6.14)	0.9633
Log CRP
Baseline	0.41 (1.03)	0.35 (0.88)	0.80 (0.94)	0.70 (1.03)
Mean differences	–0.10 (1.09)	0.01 (0.99)	–0.21 (1.00)	0.01 (1.00)	0.3737

Mean Δ for blood pressure: mean difference during the trial from baseline.

Mean Δ for cholesterol: mean difference between end of study and baseline values.

Difference between active and placebo arms within groups significant (P < 0.05).

Apo: apolipoprotein; DBP: diastolic blood pressure; HCTZ: hydrochlorothiazide; HDL: high-density lipoprotein; CRP: C-reactive protein; LDL: low-density lipoprotein; SBP: systolic blood pressure; SD: standard deviation.

Table 8.

Blood pressure response to candesartan/HCTZ.

	Non-Asian		Asian
			Chinese		Other Asians
	Candesartan/ HCTZ 3238	Placebo 3224	Candesartan/ HCTZ 1844	Placebo 1847	Candesartan/ HCTZ 1274	Placebo 1276
SBP
Baseline SBP, mmHg (SD)	136.6 (15.1)	136.0 (14.9)	140.8 (12.6)	140.9 (12.6)	138.5 (16.0)	138.3 (16.6)
SBP: mean difference post run in, mmHg (SD)	–10.1 (13.33)	–2.4 (12.9)	–10.3 (11.00)	–8.2 (11.4)	–9.4 (14.5)	–2.1 (13.6)
SBP: average change between candesartan/ HCTZ and placebo, mmHg (SD)	7.70 (13.33)		2.15 (11.17)		7.28 (14.05)
DBP
Baseline DBP, mmHg (SD)	81.4 (9.6)	81.3 (9.5)	82.3 (8.6)	82.2 (8.8)	82.8 (9.8)	82.5 (9.3)
DBP: mean difference post run in, mmHg (SD)	–5.9 (8.4)	–1.9 (7.7)	–5.4 (7.3)	–4.3 (7.8)	–5.5 (9.0)	–2.1 (7.8)
DBP: average change between candesartan/ HCTZ and placebo, mmHg (SD)	3.93 (8.06)		1.09 (7.59)		3.38 (8.45)

DBP: diastolic blood pressure; HCTZ: hydrochlorothiazide; SBP: systolic blood pressure; SD: standard deviation.

Table 9.

Use of non-study blood pressure-lowering medication by random assignment to candesartan/HCTZ and placebo.

	Non-Asian				Chinese
	Cand/HCTZ 3238		Placebo 3226		Cand/HCTZ 1844		Placebo 1847		Cand/HCTZ 1274		Placebo 1276
	N ^a	NS Meds	N ^a	NS Meds	N ^a	NS Meds	N ^a	NS Meds	N ^a	NS Meds	N ^a	NS Meds
Randomisation	3238	902 (27.9%)	3226	933 (28.9%)	1844	243 (13.2%)	1847	241 (13.0%)	1274	243 (19.1%)	1276	221 (17.3%)
24 Months	3089	724 (23.4%)	3073	1006 (32.7%)	1827	135 (7.4%)	1829	245 (13.4%)	1212	152 (12.5%)	1221	263 (21.5%)
EUF	2926	726 (24.3%)	2903	1024 (35.3%)	1777	134 (7.5%)	1775	309 (17.4%)	1162	208 (17.9%)	1176	355 (30.2%)

Number of subjects with evaluable data.

Includes any of alpha-blocker, calcium antagonist, non-thiazide diuretic or aldosterone antagonist.

HCTZ: hydrochlorothiazide; NS Meds: Non study medications.

Table 10.

Permanent discontinuation of candesartan/HCTZ and placebo by ethnicity.

	Candesartan/HCTZ		Placebo
	No. randomised	Discontinued (%)	No. randomised	Discontinued (%)
Non-Asian	3238	1075 (33.20)	3226	1183 (36.67)
Chinese	1844	104 (5.64)	1847	126 (6.82)
Other Asian	1274	373 (29.28)	1276	289 (22.65)

HCTZ: hydrochlorothiazide.

The placebo event rates were similar in Asians and non-Asians (Figure 2, Table 11). There were no significant differences between groups in treatment effects in either of the composite primary outcomes, cardiovascular death or MI. A nominally significant reduction in stroke was observed in Asians in the BP-lowering arm (HR 0.67, 95% confidence interval (CI) 0.45–0.99) but not in non-Asians (HR 1.06, 95% CI 0.65–1.73). However, the test for interaction was not significant (P = 0.146). In an analysis by tertiles of SBP, apparent harm was observed in the lowest tertile (SBP ≤131.5 mm), HR 2.72, 95% CI 1.36–5.44) among Asians which was not observed in non-Asians (HR 1.14, 95% CI 0.68–1.90). The P for interaction was significant at 0.0408 (Table 12). Despite the lower reduction in BP in Chinese, the impact on clinical events was similar in Chinese compared with others (Tables 13 and 14).

Figure 2.

Kaplan–Meir curves for co-primary outcome 2, stroke and myocardial infarction for blood pressure lowering in Asians and non-Asians.

Table 11.

Clinical outcomes.

	Asians			Non-Asians
	Candesartan/ HCTZ	Placebo	HR (95% CI)	Candesartan/ HCTZ	Placebo	HR (95% CI)	P_interaction Asian vs.
	N = 3118	N = 3123		N = 3238	N = 3226		non-Asians
BP-lowering arm
Co-primary outcome 1	128 (4.11%)	129 (4.13%)	0.99 (0.78–1.27) P = 0.965	132 (4.08%)	150 (4.65%)	0.88 (0.69–1.11) P = 0.268	0.466
Co-primary outcome 2	148 (4.75%)	142 (4.55%)	1.05 (0.83–1.32) P = 0.707	164 (5.06%)	186 (5.77%)	0.88 (0.71–1.08) P = 0.218	0.269
CV death	73 (2.34%)	74 (2.37%)	0.99 (0.72–1.37) P = 0.949	82 (2.53%)	96 (2.98%)	0.85 (0.63–1.14) P = 0.286	0.503
MI	25 (0.80%)	22 (0.70%)	1.14 (0.64–2.02) P = 0.654	27 (0.83%)	40 (1.24%)	0.67 (0.41–1.10) P = 0.111	0.170
Stroke	42 (1.35%)	63 (2.02%)	0.67 (0.45–0.99) P = 0.043	33 (1.02%)	31 (0.96%)	1.06 (0.65–1.73) P = 0.810	0.146
	Rosuvastatin	Placebo		Rosuvastatin	Placebo
	N = 3122	N = 3119		N = 3239	N = 3225
Cholesterol-lowering arm
Co-primary outcome 1	114 (3.65%)	143 (4.58%)	0.79 (0.62–1.01) P = 0.063	121 (3.74%)	161 (4.99%)	0.74 (0.58–0.94) P = 0.012	0.694
Co-primary outcome 2	127 (4.07%)	163 (5.23%)	0.77 (0.61–0.98) P = 0.030	150(4.63%)	200 (6.20%)	0.74 (0.60–0.91) P = 0.005	0.771
CV death	71 (2.27%)	76 (2.44%)	0.93 (0.67–1.29) P = 0.663	83 (2.56%)	95 (2.95%)	0.86 (0.64–1.16) P = 0.327	0.738
MI	24 (0.77%)	23 (0.74%)	1.04 (0.59–1.84) P = 0.892	21 (0.65%)	46 (1.43%)	0.45 (0.27–0.76) P = 0.003	0.033
Stroke	42 (1.35%)	63 (2.02%)	0.66 (0.45–0.98) P = 0.038	28 (0.86%)	36 (1.12%)	0.77 (0.47–1.26) P = 0.297	0.637
	Both active treatments	Double placebo		Both active treatments	Double placebo
	N = 1556	N = 1557		N = 1624	N = 1611
Combined BP and cholesterol lowering
Co-primary outcome 1	54 (3.47%)	69 (4.43%)	0.78 (0.55–1.11) P = 0.170	59 (3.63%)	88 (5.46%)	0.66 (0.47–0.91) P = 0.012	0.485
Co-primary outcome 2	64 (4.11%)	79 (5.07%)	0.81 (0.58–1.12) P = 0.201	72 (4.43%)	108 (6.70%)	0.65 (0.48–0.88) P = 0.005	0.342

BP: blood pressure; CI: confidence interval; CV: cardiovascular; HCTZ: hydrochlorothiazide; HR: hazard ratio; MI: myocardial infarction.

Table 12.

Co-primary outcome 2 by tertiles of baseline systolic blood pressure.

Co-primary outcome 2 (CV death/MI/stroke/resusc arrest/revsc/HF)
	Both active treatments		Rosuvastatin active alone		Candesartan/ HCTZ active alone		Both placebo		Candesartan/HCTZ active vs. both placebo			Both active vs. both placebo
SBP	N	Events (rate %)	N	Events (rate %)	N	Events (rate %)	N	Events (rate %)	HR	95% CI	P _interaction	HR	95% CI	P _interaction
3rd Tertile: >143.5/mean (SD) = 154.1 (8.91)
Asian	618	32 (5.18)	600	36 (6.00)	625	33 (5.28)	609	43 (7.06)	0.74	0.47–1.17		0.73	0.46–1.16
Non-Asian	453	22 (4.86)	448	34 (7.59)	460	36 (7.83)	427	43(10.07)	0.78	0.50–1.21	0.888	0.47	0.28–0.79	0.218
2nd Tertile: 131.6–143.5/mean (SD) = 137.6 (3.34)
Asian	514	23 (4.47)	536	17 (3.17)	513	22 (4.29)	518	25 (4.83)	0.88	0.50–1.56		0.93	0.53–1.64
Non-Asian	536	28 (5.22)	543	19 (3.50)	557	26 (4.67)	544	37 (6.80)	0.70	0.42–1.15	0.549	0.76	0.47–1.24	0.610
1st Tertile: ≤131.5/mean (SD) = 122.2 (7.51)
Asian	424	9 (2.12)	430	10 (2.33)	424	29 (6.84)	430	11 (2.56)	2.72	1.36–5.44		0.81	0.34–1.96
Non-Asian	635	22 (3.46)	624	25 (4.01)	597	30 (5.03)	638	28 (4.39)	1.14	0.68–1.90	0.0408	0.77	0.44–1.34	0.909

CI: confidence interval; CV: cardiovascular; HCTZ: hydrochlorothiazide; HF: heart failure; HR: hazard ratio; MI: myocardial infarction; SD: standard deviation.

Table 13.

Co-primary outcome 1 by ethnicity.

	Candesartan/HCTZ		Placebo		Co-primary outcome 1
	No. randomised	N (%)	No. randomised	N (%)	HR	95% CI	P _interaction
Overall	6356	260 (4.09)	6349	279 (4.39)	0.93	0.79–1.10
Non-Asian	3238	132 (4.08)	3226	150 (4.65)	0.88	0.69–1.11
Asian	3118	128 (4.11)	3123	129 (4.13)	0.99	0.78–1.27	0.466
South Asian	932	53 (5.69)	922	42 (4.56)	1.27	0.85–1.90
Chinese	1844	56 (3.04)	1847	66 (3.57)	0.85	0.59–1.21
Other Asian	342	19 (5.56)	354	21 (5.93)	0.92	0.50–1.72
Black African	116	14 (12.07)	109	13 (11.93)	1.02	0.48–2.18
White Caucasian	1284	44 (3.43)	1262	50 (3.96)	0.87	0.58–1.30
Latin American	1739	67 (3.85)	1757	84 (4.76)	0.80	0.58–1.10
Other	99	7 (7.07)	98	3 (3.06)	2.37	0.61–9.17	0.483

CI: confidence interval; HCTZ: hydrochlorothiazide; HR: hazard ratio.

Table 14.

Co-primary outcome 2 by ethnicity.

	Candesartan/HCTZ		Placebo		Co-primary outcome 2
	No. randomised	N (%)	No. randomised	N (%)	HR	95% CI	P _interaction
Overall	6356	312 (4.91)	6349	328 (5.17)	0.95	0.81–1.11
Non-Asian	3238	164 (5.06)	3226	186 (5.77)	0.88	0.71–1.08
Asian	3118	148 (4.75)	3123	142 (4.55)	1.05	0.83–1.32	0.269
South Asian	932	58 (6.22)	922	45 (4.88)	1.29	0.88–1.91
Chinese	1844	70 (3.80)	1847	73 (3.95)	0.96	0.69–1.33
Other Asian	342	20 (5.85)	354	24 (6.78)	0.85	0.47–1.54
Black African	116	15 (12.93)	109	13 (11.93)	1.10	0.52–2.31
White Caucasian	1284	54 (4.21)	1262	69 (5.47)	0.77	0.54–1.09
Latin American	1739	88 (5.06)	1757	100 (5.69)	0.88	0.66–1.17
Other	99	7 (7.07)	98	4 (4.08)	1.76	0.51–6.02	0.494

CI: confidence interval; HCTZ: hydrochlorothiazide; HR: hazard ratio.

Cholesterol lowering

Rosuvastatin lowered LDL-cholesterol significantly in both non-Asians and Asians but the reduction was less in Asians than in non-Asians (0.49 mmol/L (19.1 mg/dL) vs. 0.95 mmol/L (36.7 mg/dL), respectively, P_interaction < 0.0004, Table 7 and Figure 3). This reduced effect was observed in both Chinese and non-Chinese Asians although the number of evaluable samples in the latter was small (Tables 15, 16 and 17). There was a significant reduction in triglycerides in non-Asians which was not seen in Asians (–0.29 mmol/L (−25.6 mg/dL) non-Asians vs. –0.02 mmol/L (−2.14 mg/dL) Asians, respectively, P_interaction = 0.0096). Rosuvastatin lowered high sensitivity C-reactive protein in non-Asians and Asians to the same extent (–4.67 nmol/L (–0.49 mg/L) and –5.05 nmol/L (−0.53 mg/L), respectively).

Figure 3.

Levels of low-density lipoprotein cholesterol, apo-lipoprotein B and C-reactive protein by rosuvastatin and placebo in Asians and non-Asians.

Table 15.

LDL-cholesterol changes from baseline to study end rosuvastatin versus placebo Chinese.

	Rosuvastatin active		Rosuvastatin placebo		Difference (placebo-active)		P value
LDL	N	Mean (SD)	N	Mean (SD)	Mean	95% CI	P value
Baseline	396	128.3 (40.44)	387	129.6 (38.01)
Study end	396	105.7 (39.20)	387	121.6 (34.52)	15.82	10.64–21.01	<0.0001
Study end –baseline	396	–22.51 (45.31)	387	–8.04 (39.49)	14.47	6.50–20.44	<0.0001
% reduction (study end – baseline)	396	12.25 (38.06)	387	0.47 (38.33)	12.72	7.36–18.08	<0.0001

CI: confidence interval; LDL: low-density lipoprotein.

Table 16.

LDL-cholesterol changes from baseline to study end rosuvastatin versus placebo non-Chinese Asians.

	Rosuvastatin active		Rosuvastatin placebo		Difference (placebo-active)		P value
LDL	N	Mean (SD)	N	Mean (SD)	Mean	95% CI	P value
Baseline	61	116.9 (34.23)	70	113.2 (38.69)
Study end	61	98.2 (36.77)	70	104.6 (42.47)	6.41	–7.42–20.24	0.3610
Study end –baseline	61	–18.8 (52.92)	70	–8.67 (55.25)	10.13	–8.65–28.90	0.2879
% reduction (study end – baseline)	61	7.32 (47.48)	70	3.19 (56.01)	10.52	–7.58–28.61	0.2523

CI: confidence interval; LDL: low-density lipoprotein.

Table 17.

LDL-cholesterol changes from baseline to study end rosuvastatin versus placebo non-Asians.

		Rosuvastatin active		Rosuvastatin placebo		Difference (placebo-active)		P value
LDL		N	Mean (SD)	N	Mean (SD)	Mean	95% CI	P value
	Baseline	331	130.1 (30.93)	334	129.1 (30.91)
	Study end	331	90.4 (38.06)	334	122.8 (34.190)	32.48	26.97–37.99	<0.0001
	Study end –Baseline	331	–39.75 (40.50)	334	–6.26 (34.20)	33.49	27.78–39.19	<0.0001
	% reduction (study end – baseline)	331	28.58 (30.51)	334	2.03 (27.15)	26.55	22.15–30.95	<0.0001

CI: confidence interval; LDL: low-density lipoprotein.

Among Asians there was a significant 21% and 23% reduction in co-primary outcomes 1 and 2, respectively (Table 11 and Figure 4). This was similar to reductions seen in non-Asians (26% reduction for both primary outcomes). Both Asians and non-Asians had a significant reduction in stroke – by 34% and by 23%, respectively (P_interaction = 0.637). A significant 55% reduction in MI was seen only in non-Asians (P_interaction = 0.033) but not in Asians.

Figure 4.

Kaplan–Meir curves for co-primary outcome 2, stroke and myocardial infarction for cholesterol lowering in Asians and non-Asians.

Combined BP and cholesterol lowering

Asians and non-Asians receiving both active BP and cholesterol-lowering medication had similar and significant reductions in CVD events compared with those receiving double placebo (Table 11 and Figure 5).

Figure 5.

Kaplan–Meir curves for co-primary outcome 2, stroke and myocardial infarction for combined cholesterol and blood pressure lowering in Asians and non-Asians.

Safety

Dizziness/light-headedness was more frequent in the active candesartan-hydrochlorothiazide arm in both Asians and non-Asians. Among Asians these symptoms were a reason for permanent discontinuation of study medication in 3.0% in the active arm and 1.9% in the placebo arm, P = 0.0086 (Table 18). In non-Asians the corresponding numbers were 3.9% and 2.2%, P < 0.0001. The incidence of syncope was numerically but not significantly higher with BP lowering, but the rates of syncope were low among both Asians (four active, zero placebo) and non-Asians (eight active, five placebo).

Table 18.

Permanent discontinuation of medication and adverse effects.

Outcome	Asians	Non-Asians
BP-lowering arm
Any reason – placebo	415/3123 (13.3%)	1183/3226 (36.7%)
Any reason – active	477/3118 (15.3%)	1075/3238 (33.2%)
P value	0.025	0.004
Dizziness or hypotension – placebo	60/3123 (1.9%)	70/3226 (2.2%)
Dizziness or hypotension – active	92/3118 (3.0%)	125/3238 (3.9%)
P value	0.009	<0.0001
Cholesterol-lowering arm
Any reason – placebo	440 (14.1%)	1224 (38.0%)
Any reason – active	428 (13.7%)	1082 (33.4%)
P value	0.661	0.0001
Muscle pain or weakness – placebo	16 (0.5%)	60 (1.9%)
Muscle pain or weakness – active	18 (0.6%)	65 (2.0%)
P value	0.864	0.718

BP: blood pressure.

Muscle-related symptoms leading to permanent discontinuation of rosuvastatin were similar between the active and placebo groups among Asians and non-Asians (Table 18). However, temporary discontinuation for muscle-related symptoms was more frequent with rosuvastatin in both Asians (1.0% rosuvastatin vs. 0.5% placebo, P = 0.035) and non-Asians (1.7% rosuvastatin vs. 0.8% placebo, P = 0.002). Rhabdomyolysis or myopathy was uncommon in both Asians (zero rosuvastatin, one placebo) and non-Asians (two rosuvastatin, zero placebo).

A significant 33% excess in new diabetes was noted with rosuvastatin in non-Asians (4.70% rosuvastatin vs. 3.52% placebo, P = 0.025), but not in Asians (3.02% rosuvastatin vs. 4.04% placebo, P_interaction = 0.002). The main study reported a higher incidence of cataract surgery in participants on active rosuvastatin. This was not observed in Asians – 1.66% with rosuvastatin versus 1.49% with placebo. In contrast, there was an increase in cataract surgery among non-Asians – 5.02% rosuvastatin versus 3.79% placebo, P = 0.02. However, the test for interaction was not significant (P = 0.4291).

Discussion

Overall, the impact of both the trial interventions (BP lowering and lipid lowering) on clinical events were similar in Asians and non-Asians. There was a neutral effect with BP lowering but a significant reduction of both primary outcomes of similar magnitude with lowering of LDL-cholesterol. The study medications were well tolerated by both groups, with no excess in muscle-related symptoms in Asians with rosuvastatin. An excess of diabetes and cataract surgery was seen only in non-Asians, but not in Asians. Therefore, contrary to common belief, in HOPE 3, rosuvastatin 10 mg daily was at least as well tolerated in Asians as in non-Asians.

Despite better adherence to medication the degree of BP and LDL lowering was less in Asians, with both interventions compared with placebo. SBP was lowered by 4.24 mmHg in Asians compared with 7.7 mmHg in non-Asians. The reduction in SBP among non-Chinese Asians was similar to that seen in non-Asians (−7.28 mmHg), but the reduction in SBP among Chinese was substantially smaller (–2.15 mmHg). Given that adherence to randomised treatments was higher among Asians (including Chinese participants) compared with non-Asians and the use of non-study medications for lowering BP was similar in Asians and non-Asians, we do not have a coherent explanation for the lesser effect of candesartan plus hydrochlorothiazide in lowering BP among Chinese. While neither group showed significant reductions overall in primary outcomes with BP lowering, Asians in the active BP-lowering arm did have a significant 33% reduction in stroke, which was not evident in non-Asians, although the test for interaction was not nominally significant. This could be due to the higher incidence of stroke among Asians (2.02% in Asians compared with 0.96% in non-Asians in the placebo arm) or the higher initial BP in Asians. The difference in impact on strokes between Asians and non-Asians could also be due to chance. A prespecified analysis of the benefits of BP lowering by thirds of SBP in the main HOPE trial showed a neutral outcome in subjects in the lowest third of SBP.⁸ In the present analysis, Asian subjects in the lowest third of SBP (mean SBP 122.2 mmHg) showed apparent harm, while in non-Asians there was a neutral effect (Table 12). This could suggest that in Asians, BP lowering in those with SBP below 130 mmHg should be done with caution, if at all.

The impact of adherence or otherwise to a healthy lifestyle has not been addressed in this analysis. Such an analysis in the entire HOPE 3 study population¹² assessed the effect of drug interactions in subgroups by adherence to two or more of four healthy lifestyle factors – (non-smoking (never smokers and those who had quit more than a year earlier) versus current smoking; moderate or vigorous physical activities versus less activity; optimal waist/hip ratio (≤0.89 for men and ≤0.84 for women) versus a higher ratio and a healthy versus unhealthy diet, assessed using a 9-point Mediterranean diet score) with those with less than two. This analysis reported that those who had two or more of these factors had overall lower rates of cardiovascular events (HR 0.85, 95% CI 0.73–1.00). Rosuvastatin was beneficial to both groups – two or more healthy lifestyle factors (HR 0.85, 95% CI 0.62–0.90) and less than two healthy lifestyle factors (HR 0.79, 95% CI 0.61–1.01). On the other hand, BP reduction with candesartan/hydrochlorothiazide reduced outcomes only in those with less healthy lifestyles – two or more healthy lifestyle factors (HR 1.00, 95% CI 0.83–1.20) and less than two healthy lifestyle factors (HR 0.78, 95% CI 0.61–1.00). This could be due to the lower event rate and lower baseline BP in those with healthier lifestyles.

Previous reports have suggested that Asians have greater cholesterol lowering and higher plasma exposure to rosuvastatin in pharmacokinetic studies compared with other ethnic groups.^11,13,14 By contrast, in HOPE 3 while there was a significant reduction in LDL-cholesterol in both Asians and non-Asians, the magnitude of reduction in LDL-cholesterol was less in the Asians despite better adherence to medication. It should be noted that while baseline cholesterol measurements were done in all subjects, repeat measurement was done only in a sample of the population which was selected randomly by the project office (Table 1). This approach minimises the potential for biases in the selection of participants in whom lipid measurements were obtained. Despite the apparent lesser degree of reduction in LDL-cholesterol, the reductions in primary outcome events were surprisingly similar in Asians and non-Asians. In the cholesterol-lowering arm, there was a significant 34% and 23% reduction in stroke among Asians and non-Asians, respectively. While non-Asians showed a significant 55% reduction in MI, this was not apparent in Asians. These differences may be due to chance because of the small number of events in this subgroup analysis or may suggest different benefits for MI versus stroke in Asians compared with non-Asians. This should be systematically examined across all trials comparing the effects of BP lowering and cholesterol lowering in Asians and non-Asians.

Rosuvastatin 10 mg was well tolerated, with no difference between the active and placebo arms either in Asians or non-Asians in permanent discontinuation due to muscle pain or weakness (Table 18). Only three cases of rhabdomyolysis/myopathy were reported in the entire trial. There was no increase in serious muscle-related symptoms among Asians, including 3691Chinese. Pharmacokinetic studies have reported higher plasma levels of statins in Asians,^13,14 and although no specific increase in muscle toxicity has been reported there has been an impression that lower doses of statins should be used in Asians.¹¹ In the HPS-THRIVE study¹⁵ of laropripant and niacin versus placebo on a background treatment of simvastatin 40 mg in 25,672 patients including 10,932 from China, the incidence of any myopathy in the control arm was low but still higher among the Chinese than among Europeans – 0.13%/year in China versus 0.04%/year in Europe, P = 0.001. Most reports of myopathy are associated with higher doses of simvastatin, and possibly the moderate dose of rosuvastatin we used avoided any excess of adverse effects in both Asians and non-Asians. There have been reports of increased rates of new-onset diabetes mellitus associated with the use of statins. In the JUPITER trial¹⁶ a significant 25% increase in new-onset diabetes mellitus was reported in the statin arm among 17,802 randomly assigned patients. However, 20 mg of rosuvastatin was used in JUPITER compared with 10 mg in HOPE 3 and few Asians were included. In HOPE 3 we did not observe any increase in new diabetes mellitus in Asians, but there was a 33% excess in non-Asians (which is consistent with JUPITER). The reasons for this are unclear and could include differences in BMI or genetic differences in the 3-hydroxy-3-methylglutaryl coenzyme A (HMGCoA) reductase enzyme (which has been linked to diabetes).¹⁷ Examining the effects of specific statins at comparable doses on LDL-cholesterol, triglycerides and glucose levels in other studies in Asians and non-Asians is necessary to inform whether the differences we observed in HOPE 3 are real or due to chance.

Conclusion

The results of HOPE 3 indicate that the use of a fixed dose of rosuvastatin 10 mg daily was effective and safe in the primary prevention of CVD in individuals aged over 55 years (women over 60 years) at intermediate risk (Figure 4). Those with SBP over 140 mmHg had incremental benefit with additional BP lowering with fixed dose candesartan 16 mg/hydrochlorothiazide 12.5 mg. With over 6000 Asian participants, the present analysis shows reliably that Asians had similar benefits and at least similar safety with statins and BP lowering compared with non-Asians. However, the lesser degree of BP lowering seen with candesartan plus hydrochlorothiazide among Chinese suggests that future studies should be large enough to explore the consistency of effects in Asians (who constitute a very large part of the world’s population) and in other ethnic groups.

Strategies to address the problem of the surging CVD burden and its risk factors successfully have been hampered by factors such as affordability, access and compliance.¹⁸ The benefit seen in HOPE 3 was achieved using fixed doses of medications without frequent visits for BP, cholesterol measurements or safety blood tests, thereby potentially saving significant costs to the healthcare system and making this strategy easily applicable in various countries and healthcare systems. Such a strategy facilitates large-scale community level interventions with statins and BP-lowering drugs at low or medium doses, to prevent CVD using trained non-physician health workers with a clear protocol and under adequate supervision. Such a model is now being tested in the HOPE 4 trial in Malaysia and in Colombia.^18,19 The success of such trials could open the way to the development of polypills for a subset of people such as those with diabetes mellitus.²⁰

Author contribution

PP and SY drafted and finalised the paper. SY, JB and EL conceptualised and designed the HOPE 3 study. HJ, PP and SY analysed the data for this paper. PP, AD, JZ and LL contributed to the acquisition of data. AD, JZ, LL, DK, JB and EL critically revised the manuscript. All authors gave final approval and agree to be accountable for all aspects of the work ensuring accuracy and integrity.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The HOPE 3 Study was supported by a grant (IR2-91038) from the Canadian Institutes of Health Research and by an unrestricted grant from AstraZeneca. The authors received no additional financial support for the authorship, analysis or publication of this paper.

Supplemental Material

Supplemental material for Impact of blood pressure lowering, cholesterol lowering and their combination in Asians and non-Asians in those without cardiovascular disease: an analysis of the HOPE 3 study

Supplemental Material for Impact of blood pressure lowering, cholesterol lowering and their combination in Asians and non-Asians in those without cardiovascular disease: an analysis of the HOPE 3 study by Prem Pais, Hyejung Jung, Antonio Dans, Jun Zhu, Lisheng Liu, Deepak Kamath, Jackie Bosch, Eva Lonn and Salim Yusuf in European Journal of Preventive Cardiology

References

World Health Organization. Global Action Plan for the Prevention and Control of NCDs 2013–2020. Geneva, Switzerland: World Health Organization, Department of Noncommunicable Diseases and Mental Health. 2013: 7. www.who.int/nmh/events/ncd_action_plan/en/ (accessed 27 November 2018).

Pearson

Blair

Daniels

et al.

AHA Guidelines for primary prevention of cardiovascular disease and stroke: 2002 update. Circulation 2002; 106: 388–391.

Meschia

Bushnell

Boden-Albala

et al.

Guidelines for the primary prevention of stroke: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2014; 45: 3754–3832.

Wald

Law

. A strategy to reduce cardiovascular disease by more than 80%. BMJ 2003; 326: 1419–1424.

Yusuf

Pais

Afzal

et al.

The Indian Polycap Study (TIPS). Effect of a polypill (Polycap) on risk factors in middle-aged individuals without cardiovascular disease: a phase II double blind, randomized trial. Lancet 2009; 373: 1341–1351.

Rodgers

Patel

Berwanga

et al.

Pill Collaborative Group. An international randomized placebo-controlled trial of a four component combination pill (“polypill”) in people with raised cardiovascular risk. PLoS One 2011; 6: e19857–e19857.

Yusuf

Lonn

Pais

et al.

for the HOPE-3 Investigators. Blood-pressure and cholesterol lowering in persons without cardiovascular disease. N Engl J Med 2016; 374: 2032–2043.

Lonn

Bosch

López-Jaramillo

et al.

for the HOPE-3 Investigators. Blood-pressure lowering in intermediate-risk persons without cardiovascular disease. N Engl J Med 2016; 374: 2009–2020.

Yusuf

Bosch

Dagenais

et al.

for the HOPE-3 Investigators. Cholesterol lowering in intermediate-risk persons without cardiovascular disease. N Engl J Med 2016; 374: 2021–2031.

10.

Liao

. Safety and efficacy of statins in Asians. Am J Cardiol 2007; 99: 410–414.

11.

Kim

McArthur

et al.

Statin safety in Chinese: a population-based study of older adults. PLoS One 2016; 11: e0150990–e0150990.

12.

Dagenais

Jung

Lonn

et al.

Effects of lipid-lowering and antihypertensive treatments in addition to healthy lifestyles in primary prevention: an analysis of the HOPE 3 trial. J Am Heart Assoc 2018; 7: e008918–e008918.

13.

Lee

Ryan

Birmingham

et al.

Rosuvastatin pharmacokinetics and pharmacogenetics in white and Asian subjects residing in the same environment. Clin Pharmacol Ther 2005; 78: 330–341.

14.

Birmingham

Bujac

Elsby

et al.

Rosuvastatin pharmacokinetics and pharmacogenetics in Caucasian and Asian subjects residing in the United States. Eur J Clin Pharmacol 2015; 71: 329–340.

15.

HPS2-THRIVE Collaborative Group. HPS2-THRIVE randomized placebo-controlled trial in 25673 high risk patients of ER niacin/laropripant: trial design, prespecified muscle and liver outcomes and reasons for stopping study treatment. Eur Heart J 2013; 34: 1279–1291.

16.

Ridker

Danielson

Fonseca

FAH

et al.

for the JUPITER Study Group. Rosuvastatin to prevent vascular events in men and women with elevated C-reactive protein. N Engl J Med 2008; 359: 2195–2207.

17.

Swerdlow

Preiss

Kuchenbaecker

et al.

HMG-coenzyme A reductase inhibition, type 2 diabetes, and bodyweight: evidence from genetic analysis and randomised trials. Lancet 2015; 385: 351–361.

18.

Castellano

Fuster

Jennings

et al.

The role for polypill for secondary prevention in ischaemic heart disease. Eur J Prev Cardiol 2017; 24: 44–51.

19.

Hamilton Health Sciences Corporation. Heart Outcomes Prevention and Evaluation 4 (HOPE-4). Retrieved from https://clinicaltrials.gov/ct2/show/NCT01826019 (2014, accessed 7 December 2018). Clinical Trials.gov identifier: NCT01826019.

20.

Janssen

Visseren

de Boer

et al.

on behalf of the SMART Study Group. Combined use of polypill components in patients with type 2 diabetes mellitus. Eur J Prev Cardiol 2018; 25: 1523–1531.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.39 MB